Tlag the Use of Non-Explosivo Mial4res Vr Hydrogen and Oxygen

TAMU - T -79-001 c. 2 tlag IpPV r iii< The Use of Non-Explosivo MIAL4resvr Hydrogen and Oxygen For Diving

WILLIAM P. FIFE TAMU-SG-79-201 Hyperbaric Laboratory January 1979 THE USE OF NON-EXPLOSIVE MIXTURES OF

HYDROGEN AND OXYGEN FOR DIVING

William P. Fi fe

Texas A&M University Hyperbaric Laboratory

January 1979

TAMU-SG-79-201

Partially supported through Institutional Grant 04-7-158-44105 to Texas A&M University by the National Oceanic and Atmospheric Administration's Office of Sea Grants Department of Commerce

and by the

U.S. Office of Naval Research Contract Number N00014-75-C-1020 $5.00

Order from:

Marine Information Service Sea Grant College Program Texas A&M University College Station, Texas 77843 ABSTRACT

The purpose of this report is to place under one cover a

summary of hydrogen-oxygen hydrox! diving to date, and a detailed

description of the techniques which have been developed to conduct such diving in this laboratory. ACKNOWLEDGEMENTS

Dr. John P. McGrath, DVM and Veterinary Pathologist evaluated the blood chemistry. Dr. Herman R. Crookshank, Ph.D. supervised the blood chemistry analyses. Dr. William R. Klemm, DVM, Ph.D. and Dr. James Verlander, Ph.D. evaluated the electroencephalography evidence. Electron micrographs were prepared by Phillip Lesko. Special appreciation is expressed to Dr. Claes Lundgren who reviewed the original manuscript and made valuable comments, and to WayneHughes, Mark Edwards, Dennis Denton, and Donald Welch who mixed much of the hydrox gas and provided technical continuity. Apprecia- tion also is expressed to Mr. Kenneth Beerwinkle who designed much of the electronics.

The animal studies in this report were supported by the U.S.

Office of Naval Research, Contract Number N00014-75-C-1020.

The human studies were supported by the Texas A&M University

Sea Grant Program Number 04-7-158-44105.

Although the development and testing of man decompression tables was not a part of the U.S. Navy-supported program, it is included here to provide a broader picture of hydrox research. TABLE OF CONTENTS

Pacae

I. History of Hydrox Diving

II. Potential Value of Hydrox Diving II-1

III. Methodology III-1

1. Types of Hydrox Diving III-1

a. Animals III-1

b. Man III-1

2. Protocols III-4

a. Dog III-4

b. Man III-9

c. Rat and Mouse II I-8

IV. Results and Discussion IV-1

1. Brain IV-2

2 ~ Temperature IV-8

3. Bends Difference Between H2 and He IV-10

4. Fertility IV-15

5. Lung and Blood Studies IV-16

V. Special Problems and Their Solutions: V-1

1. Leaks V-1

2. C02 Removal V-2

3. Heat Loss V-4

4. Ammonia Control V-5 Pacae VI. Safety Considerations: V-1

l. Explosion V-1 2. Hydrogen Embrittlement V-2

3. Flash-back Arrestors V-3

VII ' Appendices VII-1 1. Protocol for Hydrox Diving With Animals VI I-1

2. Protocol for Hydrox Diving With Man VII-5

3. Protocol for Lung Biopsies VII-8

4. Protocol for Liver Biopsies VI I-1 0

5. Potting Techniques: VII-11 THE USE OF NON-EXPLOSIVE MIXTURES

OF HYDROGENAND OXYGENFOR DEEP DIVING

HISTORICAL BACKGROUND: The effects of hydrogen-oxygenbreathing mixtures on animals initially was examinedin 1789by Lavosier ! whoplaced guinea pigs in bell jars containing a mixture of hydrogenand oxygen. The initial concentration of oxygen was "in nearly the sameproportion in volume which exists between life-giving air and nitrogen gas in the atmosphere." The animals apparently survived 8 to 10 hours. Althoughat the time it wasfelt that they succumbedto ammonia,it is probable that they died from C02buildup since it is well known that under similar experimental conditions animals will survive longer if C02 is absorbed. One hundred forty-eight years elapsed before a hydrogen- oxygenbreathing mixture again wasexamined, this time by Caseand Haldane !. These workers breathed a mixture of hydrogen, nitrogen and oxygenin a hyperbaric chamberat an ambient pressure of 10 atmospheresfor up to 6 minutes. Theyreported no adversereaction. Even at that early date, work was cited which indicated that combustibility studies already has been carried out by a Professor A. C. G. Egerton,who assured Case and Haldanethat 4%oxygen in 96%hydrogen was non-explosive. In 1943, in part because of the anticipated difficulty in obtaining helium, Arne Zetterstrom of the SwedishNavy suggestedthe use of hydrogen-oxygenbreathing mixtures as an alternative to helium-oxygen for diving !. Zetterstrom developed an ingenious methodof cracking ammoniacatalytically, resulting in a gas mixture containing 75%hydrogen, 25%nitrogen and unchangedammonia. By re- movingthe ammoniaand adding 4%oxygen he arrived at a breathable mixture containing 72%hydrogen, 24%nitrogen and 4%oxygen. He com- pressedthis mixture into standardcylinders onboardship. in 1943 he conducted 2 successful open ocean dives on this mixture to 40 and 100 meters. He also conducted 2 additional dives to 70 and 160 meters respectively on a 2-gas mixture containing 96%hydrogen and 4%oxygen !. Onthe 160-meterdive he lost his life whenone of the two winch operators raising his platform hauled him directly to the surface without observing decompressionstops. Since upon reaching the surface he still wasbreathing 4%oxygen he succumbedto hypoxia and probably air embolism. The SwedishNavy continued for a short time after that accident to study the feasibility of hydrogen-oxygendiving under the .encour- agementof EngineerCommodore Ture Zetterstorm, the father of Arne. A laboratory explosion related to testing the safe limits of hydrogen-oxygenmixtures apparentlygreatly influences the SwedishNavy's decision to discontinue further work in this area !. After the rather unfortunate experiences in Sweden,interest in the use of hydrogen-oxygenas a breathing gas lay dormantuntil 1966 when Brauer ! began to study the possible effects of hydrogen on various mammals. His work, principally on mice and monkeys, suggested that hydrogenmight have a less narcotic or convulsive effect than helium at the samepressure ! 8! 9! 0! 1! 2!, but the full scope of biological effects were not precisely determined. His work did, however, lead him to conclude that molecular hydrogen is not toxic and is less narcotic than helium. As a result of these favorable results from short exposures with small animals, Brauer and his French co-workers attempted to carry out a human dive to 300 meters. Because of a number of problems on this dive they did not shift to the use of hydrogen-oxygen.They did, however, raise further question as to the possible narcotic effect of helium which Bennett 3! and others 4! had reported occurred below 800 feet. The terms "Helium Tremors," and later "High Pressure Nervous Syndrome" HPNS!, were applied to this phenomenon. From these observations, in 1967 it appeared that at depths of 1000 feet divers might be approaching the limits at which helium could be used. In view of the clearly anesthetic properties of Xenon at one atmospheres4!, and the narcotic effect of nitrogen at 6 to 7 atmospheres,it cameas little surprise to manyworkers that helium might cause a similar response. The appearanceof the HPNScaused several workers to develop the capability to workwith hydrogen-oxygenmixtures involving animals studies in the laboratory. Edel, supported by a commercial diving companybegan to study the possible use of hydrogenas a replacement for helium in diving. The diving company financed several early man dives, beginning with hydrogen-oxygenexposures of 10 minutes at a simulated depth of 200 feet. Independently, Fife also began to develop the capability to work with hydrogen, using dogs under the support of Texas A&M University funds for Organized Research. In 1968, Fife and Edel joined efforts to conduct several extended deep saturation dives on 2 dogs using hydrogen-oxygenand helium-oxygen breathing mixtures, first at a depth of 300 FSWand then 1000 FSW. Work also continued with man dives to gradually increase hydrogen exposures. This series ended in 1969 with exposures of up to 2 hours duration 5!. Oneof us W.P.F.! was exposedto a 1$-hour 200-foot hydrogen dive, followed 8 days later by the 2-hour dive to the samedepth. No ill effects were noted from the hydrogen, although in some of the 9 shorter dives several of the divers experienced mild bends. In 1969, French workers 6! exposed a number of rabbits to a simulated depth of 944 feet 9 bars!. As a precaution they accomplished this by lowering the animal chamberfrom pontoonsto a depth of 6 meters under the ocean using external heating strips to maintain chambertemperature. A television cameraobserved the animals while their EEG's were monitored from the surface. The EEG's of these animals began to show abnormal trends within 2-3 hours and within 74 hours there was electrical silence. All animals were dead within

20 hours. Further work by Fructus and Naquet was carried out on baboons Papio papio! at simulated depths of between 300-675 meters. Of the 8 baboons used, all demonstrated abnormal EEGpatterns and 4 died. Most troubling of all was the reported presence of brain abscesses in several of the animals. These abnormal EEG's and brain abscesses will be discussed presently. Upon completion of their initial work on dogs, Edel and Fife again began to work independently, Fife emphasizing dives to 1000 FSWwith dogs, rats and mice; while Edel carried out another series of man studies. In 1976, Fife began a series of studies to develop man decompression tables for hydrogen-oxygendiving under support of the Texas A&MSea Grant Program. Working jointly with Mezzino, tables were developed and tested to a depth of 300 feet, .with bottom times of up to 30 minutes. Interim 300-foot tables of up to 45 minutes bottom time now exist. The U.S. and Swedish Navies have retained an interest in hydrox diving althoughneither at present apparentlyhas an in-househydrox capability. Therealso are informal remarksto indicate that the Soviet Union maybe examininghydrox diving but we knowof nothing definitive since the earlier work of Lazarev 7!. POTENTIAL VALUE OF HYDROX DIVING: Manydivers refuse to consider the possible use of hydrox for diving, believing that it is too dangerous. Mostdo not realize that at sufficiently low levels of oxygenthe mixture is not explosive. Although hydrogenis not nowbeing suggestedas a complete replacement for helium the merits of its use should be carefully reviewed. Firstly, it is well knownthat in somecountries there is no indiginoussource of heliumwhile in other countriesthe heliumis found in such low concentrations that its recovery and purification is very costly. Themost abundant sources of helium are found in the United States, Africa, Netherlands, and Canada. This dirth of helium causes concern, not only because of its natural scarcity but. because in somecountries political instability or reduced governmen- tal supportof reclamationprograms places the continuedavailabiity of helium in some doubt. In this country, although most users are not now concerned over potential shortageof helium, one estimate suggeststhat this country wastesover 1 billion cubic feet of helium per year, and that our presentrich sourceswill be exhaustedby the year 20008!. This does not meanthat this country would find itself without helium, but that it would be necessaryto revert to sources of low concentrations, resulting in the need for more expensive methods of extraction. Based on the laws of supply and demand,it should be expectedthat the availability will decline andthe cost will rise considerably aboveits present level. It would seemdesirable to search for another breathing gas. In the past, the cost of hydrogen-oxygengas mixtures has been about twice the cost of commercially available heliox, and, at this time to our knowledge is being mixed by only one commercial company. Thus, from the stand-point of cost it would not be economically fea- sible if procurred in this way. However, the hydrox mixing technique newly developed in our laboratory appears to be both safe and brings the cost below that of heliox. We now can produce 3% hydrox for less

than $20 per cylinder at current prices. Secondly, hydrogen has not yet been studied sufficiently to determine if it can improve the speed or safety of decompression. It seems clear that with dogs, decompression from 1000-foot saturation dives is different with hydrogen than with helium. Optimum hydrox decompression tables have not yet been worked out, but it would seem intuitively possible that the judicious use of another gas in the decompression profile could capitalize on the potential for a higher inert gas gradient. It should be noted here that none of the animal decompressiontables employedpure oxygen. The addition of oxygen in these decompression profiles would enhance the rates of decompression. It should be noted that in commercial operations a slight increase or decrease in the decompression rate after extended saturation probably would have little economic significance and would provide little incen- tive for switching to hydrox. A third possible value of hydrox is that it has a lower viscosity and density than heliox. This becomes significant at greater depths as demonstrated by Dougherty 9!. His work showed that the maximum breathing capacity of a diver at 200 feet breathing hydrox is improved approximately 40%over heliox, and 141%over compressedair. While a diver at shallow depth 00 to 600 f.w! performing at a modest work level may not find the 40% improvement parti;cularly important, it may become very important at high work levels. Further, as the diver moves to greater depths it seems clear that the increased breathing resistance will become work limiting. The precise depth at which this would become serious is not known. Lambersten and his workers 0! exposed divers to a simulated depth of 1200 feet breathing a mixture of helium, neon, and oxygen in which the breathing resistance was equivalent to breathing helium-oxygen at a depth of 5000 feet. This would suggest that man may be able to perform at low to medium work levels at these depths using heliox. However, at these depths the 40% increase in breathing capacity afforded by hydrox probably would become important. It seems clear that breathing resistance will limit the maximum diving depth long before absolute pressure causes damageat the tissue level, it may be that hydrox will be essential if man is to function effectively at deeper depths below 4000-5000 feet!, and may increase work capacity at a much shallower depth. Finally, consideration must be given to the possibility that hydrox may reduce the High Pressure Nervous Syndrome HPNS!. This was first suggested that Brauer who showedthe monkeysdid not develop HPNSon hydrox until they had reached a depth of about 1200 FSW 0!. With helium, HPNScan be expected to appear between 800-1000 FSW. Our own work failed to show gross symptoms of HPNS in either dogs or mice at a depth of 1000 FSW,although it should be noted that our compression rates were slow. METHODS

HYDROX EXPOSURE: In the laboratory, there are two general types of hydrogen diving, each calling for different techniques. In one type, the chamber is flooded with hydrox and the diver simply breathes the ambient gas mixture. In the second method, the chamber is compressed with an inert gas containing no oxygenwhile the diver breathes a hydrox mixture by mask. In this mode,since the chamberstill contains its original oxygen, it remains a normoxic mixture and the subject can breathe the ambient gas mixture in an emergencyin event of mask failure. In the second method, to reduce the build-up of hydrogen in the chamber, the exhaled gas is conductedto the outside and is dumpedinto the air through a flash-back arrestor. It should be noted that venting hydrox gas into the chamberfrom the maskdoes not create an explosive environment since the oxygen levels cannot be raised above 3% in this way. However,such a build-up would require the chamberto be flushed with nitrogen or helium containing 3%oxygen to remove the hydrogen. While the second method usually is more economical, its use on con- scious unrestrained animals is not practical since they must be intu- bated or taught to wear a mask. Thus, animal dives have used the first method while man dives so far have used the second method.

A. TECHNIQUEFOR HYDROX DIVING WITH ANIMALS, AND SATURATION DIVING WITH MAN. The detailed step-by-step procedure and the specific rationale for each step is described in Appendix1. Briefly, the procedure is to compressthe chamberwith an inert gas helium or nitrogen! to a depth of 200 FSW ata!, at which time the level of oxygenwill be reduced to 3%. Since 3% oxygen is non-explosive and at the same time is sufficient to sustain life at that depth, it then is possible to flush the chamber with 3% hydrox % oxygen in 97% hydrogen! without danger. After replacing helium and nitrogen with hydrox, the chamber can be compressed to the desired depth with pure hydrogen. During the dive, metabolized oxygen is replaced by adding 3% hydrox. During decompression, the oxygen concentration is raised to the appropriate level by flushing in 3% hydrox. In this way, hydrogen never is allowed to come in contact with oxygen at an oxygen level above 3%. Upon returning to 200 FSW, the chamber usually is flushed with 3% oxygen in 97%helium heliox!. A 3% oxygen-nitrogen mixture may be used but this may result in rather sever nitrogen narcosis if it follows an extended absence of nitrogen. After removal of the hydrogen, decompressioncan proceed in the usual manner either with helium or compressed air.

B. TECHNIQUE FOR NON-SATURATEDHYDROX DIVING WITH MAN. A detailed description of the step-by-step procedure for con- ducting hydrox chamber dives with man can be found in Appendix 2. This procedure differs from that in which the chamberis flooded with hydrox since the hydrox can be delivered by mask and the exhaled gas removed by an overboard dumping system. In developing this method, an effort has been made to reduce the number of gas mixtures required, and to employ techniques compatible with those which would be used in manned operational dive. The chamber first is compressed to a depth of 100 FSW with nitrogen while the diver breathes compressed air. He then begins to breathe 10% oxygen in helium 0% heliox! by mask while the chamber is further compressed to 200 FSWwith nitrogen. Upon reaching 200 FSW ata! the breathing mixture is shifted to 3%heliox to wash out the 10% oxygen; and then shifted to 3% hydrox. The chamber then may be compressed with nitrogen to the desired depth. Decompressionis just the reverse of compression. The chamber pressure is reduced in accordancewith the decompressionschedule until arriving at 200 FSW. The breathing mixture then may be shifted to 3% heliox to wash out the hydrox and then shifted to 10% heliox after which ascent is begun. Care must be exercised to assure that the chamberoxygen levels are raised sufficiently to sustain life as the diver ascends to provide a safety back-up in event of mask failure. Decompression may proceed normally, using the same mask for oxygen breathing which often is called for at a depth of 40 or

60 FSW. The disadvantage of this system is that on long dives the diver must wear a mask for an extended period of time. This disadvantage is overcomeif the chamberis flooded with hydrox as in the method first described. The use of the overboard dump system is further described in the discussion of safety precaution. METHODOLOGY I I I-4.

EXPERIMENTAL PROTOCOL: This work has encompassed the use of 4 different species:

1. Dog

2. Mouse

3. Rat

4. Man

A. Protocol: Do Most of the major deep dive studies in our laboratory were conducted on the mongrel dog. These dogs were chosen for their good disposition and were held in the laboratory long enough for laboratory personnel to become familiar with each personality. With two exceptions all dogs were in excellent health at the time of the dive. One dog Lurch! was found later to have heart worms but responded to treatment. Another dog Zithy! had cellulitis in one leg which was missed on her pre-dive physical examination. Two dogs Pattycake and Ginger! were about 1 month pregnant at the time of their dives. An effort was made to avoid as much medication as possible. Cortisone especially was avoided because of its possible association with aseptic bone necrosis. All dogs did, however, receive regular rabies and distemper innoculations as well as periodic worming and dipping. All animals were fed a standard dry dog food ad lib, supplemented by a daily ration of corn oil. They usually were allowed.to roam freely over a one acre field, and slept in an open shelter. Prior to each dive the subject was given a brief physical examination. This included drawing blood, and usually lung and liver biopsies, both conducted under a thorazine tranquilizer and xylocane local anesthesia. Since none of the dogs showed evidence of a pneumothoraxafter the lung biopsies, none of the dives were de-

layed from this cause. The animals were placed in the pre-heated chamber and compression begun See Appendix 2 for compression techniques!. The rate of compression was held to 20-40 feet per minute to reduce the noise level in the chamber. Further, upon arrival at each additional 100 FSW, compression was stopped for a minimum of 30 minutes to allow leak checks to be made of the chamber, and to minimize possible HPNS. All dives with dogs, except for the initial one were planned for a depth of 1000 FSW. When this depth was reached the chamber temperature was stabilized to an ambient temperature of between 32 0 0 C and 35 C with occasional adjustments made for the animals comfort. The hydrox bleed-in rate was established to match the metabolic requirements of the particular animal and to hold the chamber at oxygen levels of between 0.6% and 1.39. The effect of these oxygen levels are discussed below: The carbon dioxide scrubber motor was set at a flow rate which would hold C02 levels to no more than 0.003%. Both oxygenand C02 levels were continuously measuredby an on-line system. During the dive, the water tray occasionally was flushed to assure fresh drinking water, while the bottom of the chamber was flushed after each urination if it was noticed, or at least once each 6 hours. Since the dog usually reserved one spot for defec- ation, this was left in the chamber. It appeared to pose no serious problem. Also during the dive the attendant frequently talked to the dog via intercom. The ability of the dog to see the attendant and to hear the attendant's voice appeared to be reassuring to some animals. Several different decompression tables were used on the dogs in an effort to establish borderline profiles both for hydrox and heliox, and in this way provide a basis of comparisonbetween these two gases. Initially, an effort was madeto standardize on a decompression table which called for 15-minutes decompression stops to a depth of 800FSW, and 30-minutestops throughoutthe remainderof the ascent with longer stops at each 100-foot mark. It was found, however, that while this profile was satisfactory for heliox Figure 1! it produceddecompression sickness when used with hydrox. This is illustrated in Figure 2. Further, with hydrox, decompression sickness usually occurred in the 500 to 700-foot range. Treatment of decompression sickness depended upon the depth at which it occurred. If it occurred deeper than 100 FSW,it was found that the most satisfactory results were achieved if the chamberwas recompressed100 FSW below the depth at which it first wasobserved. However,if it occurred above 100 FSWor after reaching the surface, the animal was treated on a modified U.S. Navy Table 5. The single instance of decompression-inducedCNS bends occurring on the surface after a hydrox dive was treated on a modified U.S.N. Table 6. It will be noted in Figure 2 that in this particular instance the dog was recompressed200 feet rather than 100 feet after bends were recognized. Thesebends involved the spinal cord and resulted in loss of hind limb coordination. Becausethe animal had just awakened it could not be precisely determined at what depth the symptoms actually occurred. Since she had been seen using her hind limbs normally at the 800-foot depth, she was recompressed to 100 FSW deeper than the last asymptomatic depth, with several short pauses during recompression to evaluate symptoms. Shortly after reaching 900 FSW she regained full use of her hind limbs. The second recompression at 700 FSWwas due to a subjective judgment by the observer that the rear legs again showed weakness. This was shown to be due

to sore foot pads and not decompression sickness. As a result of these and other studies, it seems clear that recompression to depths less than 100 feet deeper than the depth of bends onset could result in only partial or temporary relief. Rarely did the 100-foot rule fail to provide adequate relief although in 2 instances the animals retained minor residual CNS symptoms which re-

ceded over the next 2 months. Since one objective of these studies was to compare the physiological effects between heliox and.hydrox diving, the decompression profile shown in Figure 4 was employed on both heliox and hydrox dives. Further, bottom times were standardized at. 48 hours for both gas mixtures. In this way, decompressionsickness was avoided in all subsequent studies. Although it is probable that when used with helium this table has a greater margin of safety against bends than for the same dive employing hydrogen, there is no evidence that this

affected the comparative physiological studies. Post-dive brain biopsies were carried out on two animals for electron microscopic evaluation. Brain necropsies were carried out on four animals in a search for brain lesions. Pre- and post-dive kidney biopsies for electron microscopic studies were carried out

on two animals. A study of core temperature during 100-foot dives was conducted. To accomplish this a small sub-carrier oscillator with FM transmit- C! 0 n O <3 C3 O O O C! C! C3 C3 O 4! OCU L3 3 3 N I Hl

p Z rrj

O N

O C3 O O 0 O O O O OEA O ln Ct J333 N I Hl lO ter was coupled with a temperature . ensor and encapsulatedin a mixture of beeswax and paraffin. The complete unit including battery was approximately 6 cm long and 3 cm in diameter and was moldedin the hands to remove sharp edges and rough surfaces. It was implanted into the peritoneal cavity one weekprior to the dive by a small mid- line abdominal incision. This type of unit operates continuously for about one month before batteries lo: e power. One such unit remained in an animal for 19 months during which time she has 2 litters. It caused no inflamatory reaction and still was serviceable when the battery was replaced. Post-hydrox breeding experiments were carried out on selected dogs, both male and female. No medication was given either to

enhance or inhibit fertility.

B. Protocol: Mouse This work was ancillary to the primary program. It involved a study to determine the possible effect which hyperbaric hydrogen may have on various types of tumors. However, since exposure techniques are applicable to other hydrox diving the work on tumor- bearing mice will be briefly considered. It should be realized that all of the mice involved in this work had malignant tumors and that 100% mortality was virtually certain. The tumors employed were:

1. Squamous cell carcinoma

2. Ehrlich's ascites tumor

3. Leukemia Tumors were mapped and animals were earmarked in accordance with standard procedures. They were placed in small chambers containing food, water, bedding, CO scrubber.. Heat was applied 2 in accordance with the techniques found in Appendix l. Since these small chambers could not be flushed with water to remove urine they were brought to the surface every 3 to 4 days. Decompression from 300 FSWwas carried out in three hours, while decompression from 1000 FSWwas carried out in 6 hours. This was considered to be conservative decompression for mice but was done

in deference to the debilitated condition of some of the animals. Except for these modifications the dive protocol was essentially

the same as that described for dogs. Tissue biopsies and necropsies were performed on some of the animals although in most instances studies consisted mainly of

survival time.

C. Protocol: Rat Studies carried out on rats were related to trying to determine

the cause of death of two dogs which unexpectedly died while on a

hydrox dive. The first group of these rats was exposed to the identical profile used on the dogs which failed to survive. In the second dive, the animals were exposed to elevated. levels of oxygen 00 mm Hg! and C02

.5%! in an effort to create the same symptoms seen in the fatal dog

dive.

D. Protocol: Man

It should be emphasized that hydrogen studies with humans were

supported by the Texas A&M University Sea Grant Program. They were completely independent of the Office of Naval Research-supported program. In this program, 15-minute and 30-minute bottom-time tables suitable for man were developed and tested. Human subjects were drawn from the director and other laboratory personnel' All were divers with experience ranging from four years to nearly 35 years. Theyranged from 23 years to 58 years of age. All were in excellent physical condition and were involved in a regular swimming exercise program. In our laboratory the developmentof mantables was carried out as follows: A hydrox/helioxdecompression table for a pig wasgenerated, using parametersalready developedand programmed.This program and this technique of using the pig as the first model has been operationally provenduring the developmentof 420FSW tables for a commercial diving company 1!. For each table the pig was subjected to its decompressionpro- file to confirm its safety for that. model. Whenit was determined to be safe for the pig, a "pig-to-man" extrapolation was carried out by computerand a mantable wasproduced Fig. 5!. Previous experienceshowed that if the pig doesnot developdecompression sickness, the extrapolation will be approximately correct for man. This table with a 15-minute bottom time was dived by an experienced diver using heliox. It then was slightly modified due to the appearance of mild bends and the same diver carried out a successful dive using heliox, followed by a successful dive using hydrox. The table next was extended to 30-minute bottom time and was dived on hydrox by one of us W.P.F.!. The appearanceof mild bends resulted in still another table Table 2!. It was dived by the same diver without difficulty, following which three other divers made dives on it. Since one of these divers developed a mild case of the bends, the table again was modified ,'Table 3!. We believe this TABLE 1 25 March 1976

Texas A&M University

Hyperbaric Laboratory Man Hydrox DecompressionTable 1

300' -l7 min B.T. ~Deth Time

300' 15 min B.T. 270 1 Time 260 1 ~Deth 6 250 1 90 7 240 1 80 10 230 3 70 11 220 3 60 13 210 4 50 200 6 Shift to 3l heliox-4 min! Shift to 10%heliox-2 min! 40 15 18 190 3 30 180 5 20 22 27 170 4 10 258 min total 160 7 Flush Chamber! 150 8 With Air! 140 9 130 11 120 15 110 16 100 8 Shift to Air! Breathing! TABLE 2

TAtIU Hydrox Decompression Table V

MAN !

30 minutes 300 feet

Time Depth Time ~De th

60' 12 02 300' 30 mig. a! 50 7. O2 280 1 5 air 260 1 4 02 250 2 40 15 02 240 2 5 air 230 2 1 02 220 30 18 O2 210 4 3 b! 5 air 200 6 02 190 2 20 13 02 180 6 5 air 6 170 17 02 160 5 10 2 02 150 9 5 air 10 140 20 02 130 12 5 air 14 120 10 02 110 18 Surface 100 9 c! 90 10 Total 320 min. 80 8 d! including ascent 70 10 time one minute travel time between stops!

a! 3-; hydrox b! shift to 3%heliox followed in 1 minute by shift to 10%heliox c! shift to 50% heliox d! shift to 50-o nitrox TABLE 3

TAMU Hydrox Decompression Table VI

MAN !

30 Minutes 300 Feet

Depth Time Time 60' 12 0 300' 30 min. a! 2 280 50 7 0 2 260 5 air 250 12 12 20 0 240 2 230 40 5 air 220 20 0 2 210 24 43 5 air 200 b! 20 0 190 2 6 2 180 30 5 air 170 20 0 2 160 69 5 5 air 150 15 0 140 10 2 130 12 20 5 0 2 120 14 5 air 110 18 20 0 100 9 c! 2 90 10 10 5 air 80 8 d! 20 0 2 70 10 Surface 5 air TOTAL 345 min

a! 3% hydrox !h! shift to 3% heliox followed in 1 minute by shift to 10% heliox c! shift to 50% heliox d! shift to 50% nitrox table now is safe to use for our further studies on hydrogen wash- out of various tissues, and will permit our technicians to dive with the dogs to a depth of 300 FSWfor a bottom time of 30 minutes if this is needed. Table 4.

Accumulated Hydrox Exposures

Duration of Subject Hydrox Exposure Total Duration Hours:Min of Dives

Dog 1140:18 1382:42

Rat-Mouse 5313:00 5378:15

Man 4:37 41:27

6457:25 6802:24

These figures do not include animal dives carried out with heliox simulating a hydrox dive. Table 5.

History of Hydrox Saturation Dives

Dog

Name Total Dive Total Time on Results Animal Depth Time Hrs:Min Hydrox Hrs:Min

Uneventful Pioneer 300 FSW 34:15 23:30

Hydrox 1000 FSW 104:40 93:10 Bends' Hematuria once post dive. Heavy ammonia in chamber.

Bends treated Zithy 1000 FSW 65:23 47:10 cellulitis present from pre-dive in- fection

Uneventful Cat 1000 FSW 85:10 68:00

Tweedle 1000 FSW 19.18 16 18 Died at depth due to hydrocarbon con- Bandit 1000 FSW 19 18 16 18 tamination of gas

Uneventful Cleo 1000 FSW 93:17 76:48

Lurch 1000 FSW 63:14 52.49 Contaminated gas. aborted dive CNS treated OK.

Beagle 1000 FSW 98: 25 83:08 Subcutaneous emphysema only. No other apparent complications.

Pattycake 1000 FSW 81:30 68:52 Uneventful

Ginger 1000 FSW 122: 17 97:17 Missed decom stop, Possible slight bend self-treated No other compoli- cations.

Five aborted dive due to failure of EEG transmitters

Sidekick 1000 FSW 117:05 97: 10 Cerebral Hypoxia ?! and convulsions at 1000 FSW. Problem corrected and dog completely well and alert on reaching surface. Table 5 continued! History of Hydrox Saturation Dives

Dog

Name Total Dive Total Time on Results Animal Depth Time Hrs:Nin Hydrox Hrs:Min

97:25 Uneventful Asher 1000 FSW 118:10

96:48 Uneventful Patton 1000 FSW 118:02

97:36 Uneventful Snap 1000 FSW 117:43

107:59 Uneventful Sidekick 1000 FSW 120:08 Table 6.

History of Heliox Saturation Dives

Dog

i%arne Time Dive Total Time on Results Animal Depth Time Hrs:Min Helium Hrs:Min

Hydrox 1000 FSW 76:10 62: 00 Bends but treated OK. Hematuria once post-dive.

Sally 1000 FSW 73: 41 58 10 CNS-treated post- dive. Recovery OK.

Snow 1000 FSW 73:41 58.10 Muscular weakness Recovery OK.

Benjamin 1000 FSW 51. 15 47:48 Uneventful

Lady 1000 FSW 51.15 47:48 Uneventful

Rover 1000 FSW 17:43 17:43 Fatal at depth due to contaminated Red 1000 FSW 17:43 17:43 gas. Died within J. min of each other.

Beagle 1000 FSW 93:04 54: 57 Mild Bends, treated OK.

Sidekick 1000 FSW 93:04 55:10 Uneventful

Spot 1000 FSW 92:26 54.41 Uneventful

Patton 1000 FSW 92:45 54:52 Uneventful

Tasha 1000 FSW 93:50 55 16 Uneventful

826:37 584:18 Results and Discussion

This laboratory now has accumulated over 6 $50 hours of actual hydroxexposure to manand animals Table 4!. A total of 16 dogshave been involved in hydroxdives Table 5! while 12dogs have been involved in heliox dives Table 6!. Most of these dives with dogswere saturation dives to a depth of 1000FSW. There- sults of these efforts now will be presented and discussed:

The earlier findings of Michaud, et al 6, 24! are puzzling. Theseworkers have reported the appearanceof brain abcesses,ab- normalEEG changes, EKG abnormalities and sometimesdeath within less than six hours after the beginning of hydrox exposureat 7 ata pressure. Wehave so far failed to observe these symptomsin our animals or in ourselves. As has been reported elsewhere, of the dogswhich have been exposed to hydrox,we have lost only two during or after hydroxexposure Table 5!. Our studies of these two deathssubsequently revealed that they weredue to the presenceof toxic hydrocarbonsin the breathinggas. Therewere no indications that hydrogencontributed to these deaths. Except for these two deaths, the only medical problemsin the hydrox dogs have been related to bends during decompressionor to what may have been cerebral hypoxiawhile at depth. It wouldappear that our animals received a sufficient hydrogen exposure to produce these symptoms since they wereexposed to this gas continuously for as long as 97 hours. Brain biopsies for electron microscopic studies were carried out on two animals, and four brains were removedand sectioned. IV-2

No lesions or other gross abnormalities were noted which resembled the earlier findings of Michaud.

BRAIN: Only four animals have developed alarming nerological symptoms while at a simulated depth of 1000 FSWon hydrox. Two diving together and one diving alone apparently were subjected to gas contaminated with volatile hydrocarbons. These three animals demonstrated what.appeared to be a CNSabnormality but only after remaining at 1000FSW for from three to twelve hours. The two diving together died quickly after the first symptomsappeared. The third survived. The symptoms first appeared as general apprehensiveness and tachypneafollowed by gradual loss of equilibrium and convulsions. In two instances this was followed quickly by a spastic- type paralysis of the forelimbs and the inability to stand erect or hold its head level. In one instance this progressed to an extensor-type muscle rigidity, opisthotonous, and the inability to stand or walk. It is, perhaps, instructive to consider the 4th instance in some detail since it apparently was not related to contaminated gas. Further, the question of central nervous systemabnormalities has been raised by other workers. This animal had been instrumented with an implanted EEGtransmitter. Bipolar EEGleads were implanted in the skull in positions F-l, 0-1, and C-Z with C-Z the common electrode as suggested by Klemm 6! Figure 5!. It, thus, was possible to obtain continuous EEG's by radio telemetry in the unrestrained animals Using the pre-dive EEGas control Figure 6!, these two channels showed no apparent abnormal brain electrical activity for the first 22 hours of exposureto hydrox, 12 hours of which were at a depth of 1000 FSW Figure 7!. FIGURE 5

Prop<>serielectrode location;>»cl identification scl>e>ne. Fronz lt. Klen>m,"Atte>npts tu Standarclize Veterinary Electr<>encephah>iericr»> fo»mal »f Vcteri»nryResearch, 29:189' 1900, I'; >S.>!. IV-3

Thirty-four hours after the beginning of hydrox exposure the dog developedopisthotonous, tachypnea, apprehensivenessfollowed by convulsions. During and immediately after the convulsions, it was, of course, impossible to record useful brain electrical activity due to the presenceof overriding electrical muscle activity. Since the onset of convulsions apparently developed over an approximately one-hour period and did not coincide with changeof the gas source, pressure, or C02 levels, it was felt at the time that it possibly could be due to hypoxia, even though the oxygen level was 0.7% .6% equals a P02 of 160 mmHg at 1000FSW! ~ For that reason when the convulsions appeared the oxygen level was quickly raised to 1.5% PO = 400 mmHg! and decompressionwas be- 2 gun. Upon reaching 890 FSW, the animal showed complete recovery from all postural symptomsalthough the EEGcontinued to show the loss of beta waves. As a result of this improvements, the animal was returned to 1000 feet for another 27 hours. This recompression created no adverse effects either behaviorally or in the EEG. During the subsequent sojourn at 1000 FSW,the EEGshowed a gradual but continuous return toward normal. It did, however, reveal a loss of the superimposedbeta band on the delta and theta waves. This activity is somewhatsimilar to that observedin humanssuffering from

HPNS During the following decompression, this animal continued to show improvementin its EEGpattern until reaching a depth of 610 FSW. At that time, the inspired oxygen levels again were reduced due to operator error. Although the dog did not convulse, it IV-4

appeared to hold its head in an abnormally high position as though "star gazing". At that time, it again began to present a frankly abnormal EEG, containing overwhelming alpha waves. During and immediately after this episode muscle artifacts maskedthe EEGso that no valid recordings could be made. This episode quickly resolved itself as the oxygen levels were raised and within a few hours, the EEGappeared to be similar to those recorded before the dive. The last recording made before arrival at the surface could hardly be destinguished from the predive recording. The cause of these convulsions has not been confirmed. There is some circumstantial evidence to support the original theory that they were related to cerebral hypoxia even though the inspired oxygen levels were normoxic. A summaryof the evidence is as follows: a!. The chamber and thus, inspired! oxygen levels were accurately measured. A BeckmanF-3 paramagnetic oxygen analyzer in series with a Beckman Model 777 were employed to provide on-line monitoring of chamberoxygen. These units were calibrated at least once each 24 hours against standard reference gases containing the sameinert background gas and either zero or 3%oxygen. A variation at any time of more than 0.03%between these analyzers initiated immediate recalibration. We, therefore, are confident that the inspired oxygen levels were accurate to within this percentage at the

time of the convulsions. b!. Chamber carbon dioxide levels were not elevated. They were constantly measured on-line with a Beckman infrared CO2 analyzer ~ This unit was calibrated each 24 hours with standard reference gasescontaining zero CO2or 0.03%C02. Any apparentin- IV-5 stability in this unit initiated immediate recalibration. At the time of the convulsions, C02 levels were 0.001%. c. The source of oxygen and hydrogen gases had been changed after the earlier problems with contaminated gas. Further, all lines, scrubbers, and mixing units had been thoroughly purged and at least 4 other dives had been carried out since the contamination problemoccurred. There subsequentlywas found no evidence of con- taminants in the gas used in this particular dive other than the presence of water vapor. d. Perhaps the strongest evidence that these convulsions were hypoxia related was the fact that even though the animal had been subjected to hydrogen continuously for 22 hours, 12 of which were spent at a depth of 1000 FSW,the EEG's remained essentially normal and convulsions did not appear until the inspired oxygen levels had dropped to 0.7% 66.5 mmHg!. Whenconvulsions appeared assent was made to 900 FSW and oxygen was introduced into the chamber. By the time oxygen levels were raised to 1.5-. 00 mmHg! convulsions had ceased and the EEG lost some of its abnormal char- acteristics. In fact, the animal showed such marked EEG and behavioral improvement that it was returned to 1000 FSWfor an additional 27 hours of bottom time. Finally, after raising the oxygen levels, the improvement in EEG continued until 46 hours later during decompression. Upon reaching 610 FSWthe inspired oxygen levels again were al.lowed to decrease to 133 mmHg again due to operator error and the animal again presented both EEGand behavioral symptomsof inpending convulsions "star gazing" with head elevated and a fixed stare. These symptomsdisappeared quickly when oxygen levels again were raised to 380 mm Hg. IV-6

We are unable to identify any variable besides the oxygen levels which seemed to be temporarily related to the convulsions described in this dive. This suggested that it was important to carry out an EEG study under varying degrees of tissue hypoxia at 1000FSW, both with heliox and hydrox in an effort to quantify the level at which hypoxia in the presenceof both hydrox and heliox begins to produce abnormal EEG's. Preliminary studies to examine this problem were carried out in 5 anesthetized dogs. The use of the anesthetized dog for EEG clinical evaluation was suggested by Klemm 6! who showed that the EEGof the anesthetized dog was highly reproduceable and was quite free from manyuncontrollable external factors which plague the EEGof the awake dog. These records still are undergoing analysis and interpretation. Recordingsfrom the anesthetized dog results in a quite different EEGthan that found in the awake animal. It seems probable from our early results that HPNSmay be suppressedin the anesthetized dog. Further, while EEG'sfrom anesthetized dogs reveal chronic cerebral abnormalities resulting from disease, we are presently uncertain if many of the subtle, reversible chanqes can be seen in the anesthetized animal. We also believe it is important to measure cerebral tissue 02. Wenow have this capability andpropose to measureit at 1000

FSW on hydrox. High Pressure Nervous Syndrome As indicated earlier, Brauer suggested that the HPNS did not appear during hydrox exposureuntil the subject reached a similated depth of 1200 FSW0!. Brauer, Bennett 3! and others noted that with helium EEG changes were observed below about 600 FSW. IV-7

No clear evidence of HPNS was seen during any heliox dives.

The few instances in which animals appeared to shiver while at 1000 FSW were attributed to being cold. This shivering could be

controlled by raising chamber temperatures No behavioral symptoms appeared which could be attributed to HPNS in any hydrox dives. However, in the convulsing dive described above Figures 8, 9, 10, and 11!, immediately after con-

vulsions ceased and for a short time thereafter, the EEG pattern

had elements which resembled HPNS similar to that described by

Bennett. It is not surprising that behavioral symptoms of HPNS were not observed in our dives because of the slow rate of compression. In early dives compression to 1000 FSWoccupied 6 hours. In all of the dives in which the EEG was monitored, compression took 7.5

hours. We have conducted no studies of the effect of varying rates of compression on the EEG in the dog. This work, therefore, does not provide insight as to the resistance or susceptibility of a diver to HPNS while breathing hydrox. Recognizing that HPNS can be explained on the basis of a membrane effect, we cannot escape the question as to whether HPNS could contain an element

attributable to cerebral hypoxia. We know of no definitive studies in which cerebral tissue Q or C02has been directly measuredduring the presenceof HPNS. We have been struck by the fact that the EEG of the animal just described did not show a HPNS-type pattern during compression or for the first 34 hours of hydrox exposure. It was not. until after the animal apparently developed cerebral hypoxia and con- IV-8 vulsed that the HPNS-typeEEG appeared. This pattern disappeared after several hours of elevated inspired oxygen. It must be recognized that the loss of the beta wave is not specific for HPNS.On the other hand, the relationship of cerebral oxygenationand beta wavesat these greater depthsneeds to be examinedunder the controlled conditions in a dog instrumented for

EEG and cerebral tissue oxygenation.

TEMPERATURE: The importanceof body heat loss in the hydrox environment cannot be ignored. Webb2! has shownthat at a depth of 800 FSW using heliox, the heat loss through respiration alone maybe greater than can be replaced by metabolic activity even during heavy exercise. Since the specific heat of hydrogenis nearly twice that of helium, it was anticipated that this would create a problem when diving with hydrox. This can be overcomein the humandiver by raising the temperatureof the inspired breathing gas in the helmet or mask. In the case of the animal diver, whenthe chamberis flooded with hydrox, the convectiveheat losses through the skin are addedto the major heat losses through respiration. This may becomesevere if te chambertemperature is held to that usually

found in the laboratory. This is demonstrated in Figure l7 which shows the relationship betweenthe ambienttemperature of a chambercompressed with hydrox, and the core temperatureof the dog exposedto that en-

vironment. It will be noticed that below about 600 FSWthe trend in the core temperatureof the subject generally followed the trend of the ambient temperature, dropping to 36.9 C when the chambertem- 40

IJJ CL 38

K ILJ

IJJ 37 bJ K O 36

UJ 32

UJ I-

LIJ m

1200 2400 2400 2400 2400 2400 17FEB. I 18 FEB I 19 FEB. I 20 FEB. I 21 FEB. I

COMPARISON BETWEEN BODY CORE TEMPERATURE AND CHAMBER AMBIENT TEMPERATURE.

THE NUMBERS ABOVE THE CORE TEMPERATURE I'igure 17. REPRESENT CHAMBER DEPTH IN FSW. perature was lowered to 29.4'C. Above600 FSWthe core temperature appearedto be rather independentof the fluctuations of the chambertemperature, but appeared more often to roughly follow a

circadian rhythm. It appearedby visual observation that below 600 FSWthe dog was more uncomfortable when its core temperature dropped than when the same drop occurred at a more shallow depth. Although we have not yet correlated EEG with core temperature at depth, as indicated above it may be that the EEGabnormalities and subsequent death of the rabbits exposed to hydrox at 30 ata as reported by Michaud and his associates 6, ' may 24! actually have been due to loss of body heat rather than to any direct effect

of hydrox. Since all of our man dives have been in a chamber compressed with nitrogen, employing delivery of hydrox by masks, our subjects do not suffer from heat loss through the skin. Indeed, since the

chamber contains 97% nitrogen the diver sometimes has been rather

warm. The diver occasionally may remark as to the feeling of cool- ness of the inspired hydrox, but even though one of us W.P.F.! has breathed hydrox continuously for two hours at a simulated depth of 200 FSW and for a total of 92 minutes at 300 FSW there was no indi-

cation of body heat loss. Inspired hydrox usually was at a temperature of approximately 22'C. In animal and human dives using both heliox and hyrox, we are struck by the close similarity of feeling between the two. Wewho have dived on both heliox and hydrox are unable to sense any difference between the two gases. The voice differences subjectively appear to be indistinguishable; the breathing resistance subjectively IV-10 appears the same although differences can be measured 9! ! and the temperature effects appear to be almost identical. When animals are placed in a chamber and taken to depth, the ambient temperature must be raised to about the same value with both hydrox and heliox to keep the body temperature at normal levels and the animal comfortable. Although we have employed temperature transmitters implanted in the abdomenof dogs, it appears that the chamber temperature can be maintained adequately by visually observing the comfort of the animal. If it shivers or lies next to the heated wall the temperature is elevated. If it avoids the wall and pants, the temperature is lowered.

It appears that the viable temperature range of the animal narrows when it is exposed to ambient. hyperbaric hydrox or heliox. This, perhaps, should not come as a surprise since the lightly clothed swimmer or diver can detect, and be bothered by, water temperature changes of 4 C. The high specific heat of hydrox and heliox, together with the greatly increased density of these gases when compressed, increase body heat loss manyfold over that of nitrogen. This apparent narrowing of the temperature tolerance has another consequence of interest as the heating system is designed. It appears that maintaining the chamber environment at a temperature of 34'C to 35'C at a depth of 1000 FSW will not accurately assure the comfort of the animal. The animal does, however, make fine temperature adjustments by changing its position in reference to the heated wall of the chamber. Instances have been noted when although the ambient temperature remained unchanged, animals pur- posefully changed their distance from the heated wall. In one instance the external heating source on a mouse cnamber was moved from its position at the side wall to the bottom so that the bottom and both sides were heated alike. Even though the ambient temperature was not altered, all animals died within 24 hours, apparently from overheating.

Comparison of H drox and Heliox Decom ression

The present work makes it possible to study the difference between hydrox and heliox decompression. The difference first was noted by Edel and Fife in their early work on dives to 1000 FSWwith dogs' In this first study it was found that the same dog using the same decompression profile from a 1000-foot saturation dive developed decompression sickness at a depth of 425 FSWon hydrox while he did not develop a similar grade of bends with heliox until reaching 10 FSW. On the other hand, these observations were not viewed as a rigorous comparison because at the time of the first 54-day hydrox dive no effort had been made to remove urine. As a result, bacterial action accelerated by the 35'C temperature had produced an extremely high ammonialevel in the chamber. It is possible that these levels could have been detrimental to gas exchange and, perhaps, even caused pulmonary damage although there was no gross evidence of such damage. For this reason this observation was considered in- conclusive. It is, perhaps, useful to study what appear to be reproducible borderline decompressions from 1000-foot saturation dives. Such a profile is shown in Figure 18. This 66/-hour profile produced subcutaneous emphysemain only one dog while the remaining animals OJ

O M

IA w xxK 0 0 Q O Q Ld >- 0 Z Q

O OJ

CU CU

~R zO

0 OO CUOO V O O O O O<0 O c> lA .l.3 3 J Nl llld10 IV-12

Showed no decompression sickness symptoms. However, any altera-

tion which increased the rate of initial ascent by shortening the

time deeper than 800 FSW resulted in bends. This may be seen in

Figure 2, in which staged ascent with shortened stops was made

from 1000 FSW to 800 FSW; and Figure 19, in which a direct

ascent was made from 1000 FSW to 900 FSW without stops. In both

of these instances the initial ascent apparently precipitated de-

compression sickness which appeared at 700 FSW and 750 FSW respectively. Both were CNS-type bends but were successfully treated.

It also was possible to develop a borderline decompression table for a 1000 FSW saturation dive using heliox in which the concentrations of helium were identical to those with hydrox. Such a profile may be seen in Fig. 1. This table required 49 hours as compared with 66» hours for hydrox. It may be seen from Figure

20 that slightly increasing the rate of ascent by reducing the time at deeper stops resulted in decompression sickness' These symptoms appeared at a depth of 500 FSW and resulted in a temporary partial paralysis of one leg. It was resolved by treatment.

The question whicn must be asked is whether the increased tendency toward decompression sickness with hydrox is due to an increased rate of diffusion, increased solubility in fat tissue, increased solubility in aqueous tissue, or perhaps to several of the above factors. A review of the solubility properties of helium and hydrogen may be seen in the following table:

Table 4.

Sol. in* Sol. in* H 0 Olive Oil

N2 0. 0141 0.076 He 0.0095 0.017 H2 0.0190 0.057 *Values determined at 38'C. c>

n Eja

O O O O O o O 0 O O O IA 1333 NI III.

O CU

O A 'D O ill Al l338 Nl Hl J'30 IV-13

H is approximately 3.4 times more soluble in fat than is 2 helium, while it is 2 times more soluble in the aqueous fraction.

The possibility that gas movement is diffusion limited must be considered. The diffusion of a gas is inversely proportional to the square root of its moleculer weight. The following table shows this relationship for nitrogen, helium and hydrogen:

Table 5. 1 gas mo1 v'1'4 wt

N 28.0134 0.188937 He 8.0052 0 ' 35348 H2 2.0159 0.70423 It can readily been seen from this table that H2 should diffuse about 2 times faster than does helium. During decompression this increased rate of diffusion should be an asset in favor of H2 if it remains in solution. However, this property may be an additional hazard if supersaturation is exceeded since tnere is more likelihood of the coalescing of gas molecules. Further, once a bubble is formed tne increased pate of gas diffusion would result in more rapid bubble growth in the case of hydrogen. As will be seen in Table 5, because of the increased solubility of H2 in both fat andwater, moreH2 molecules are present than after an identical exposure to helium. This would tend to further aggravate the growth of bubbles after exposure to hydrogen. It may be instructive to consider the additive effects of solubility and diffusion. Although it is recognized that this relationship is a complex one, it may be simplified by the following table: IV-14

Table 6A.

Bends Factor Sol in 1 1 gas olive oil ~molwe sol x vm

0.0144 N2 0. 076 0.1889 0.006 He 0. 017 0.3535 H2 0. 057 0.7042 0.040

Table 6B.

Bends Factor 1 Sol in Sol x Wm gas H2O mol wt

N2 0.0141 0.1889 0.0027 0.0034 He 0.0095 0.3535 H2 0.0190 0. 7042 0.0134

It would seemfrom a study of the bends factor that for an identical exposureand profile, the subject maybe expected to developdecompression sickness approximately 7 timesmore readily on H than on He if fat is the critical tissue; or approximately 4 times more readily if aqueousfraction is critical. It is clear that the borderline bends tables presented here do not permit a critical conclusion to be drawn concerning the importance of aqueous and lipid compartmentsin bendssusceptibility with hydrox. How- ever, for hydrogen, lipid tissue has a "Bends Factor" of 7 over that for helium. The samefactor for aqueouscompartments is ap- proximately4. Thefact that it only wasnecessary to extendde- compressiontime by about22% for hydroxsuggests that someother factors must have a major influence on bends susceptibility with hydrox. Theobservations point to the needto studythe relationship betweenhydrogen and nitrogen since their solubilities in aqueous andlipid fractions are reversed. Anexamination of the borderline IV-15

decompressiontables for both H2 and He Figures 1-4! shows that following saturation at 1000 FSW,the H2 decompressiontables required a 22% extension in decompressiontime over those for He.

The Effects of Hydrox on Fertility: No evidence has been found that exposure to hydrox has any effect on fertility. Normal litters were born to seven dogs which had been subjected to 1000-foot hydrox dives. Two of these dogs were approximately one-monthpregnant at the time of their dives. Three of these litters were sired by dogs which also had been subjected to 1000-foot hydrox dives. Further, one litter of mice was born during a 300-foot hydrox dive. All but one of the young survived and subsequently

reached maturity. Although no sperm counts of other fertility tests were carried out, we were unable to see anything unusual in the cycle of any of the dogs, or any unusual differences in the size or sex ratio of the litters. The ratio of males to females appeared to be about the same for dogs which had not been subjected to hydrox. IV-16

Ultrastructure of the Lun Pre- and post-dive lung biopsies were carried out on a total of 25 dogsand were studied by electronmicroscopy. Standard techniqueswere used to prepareand view tissues. This included buffered glutaraldehyde fixation, osmiumtetroxide staining, methylmethacrylateimbedding and viewing by a transmissionelec- tronmicroscope. A critical evaluation of this work showedlittle remarkable. In someanimals there appeared to be a slight increase in the amountof collagen but in all instances it was felt that the conditions which precipitated this were not, related either to hydrox or heliox for two reasons: Firstly, it appearedto be present in mostpre-dive specimens. Secondly,it was felt that there was insufficient time for collagen to develop during a dive even if an appropriateinsult occurred. As a further indication that such an insult did not occur, attention is directed to the animals which had a secondseries of lung biopsies related to subsequentdives. There was no evidence that these animals suffered any long-term or late-developing ultrastructural changeswhich could be attributed to a previous dive. Special attention waspaid to the alveolar lining. TheType B cells appearedto be completelynormal, containing a normalnum- ber of osmophilic lamellar bodies, mitochondria and other struc- tures. Further, the pulmonarycapillary endotheliumappeared normal in all respects. Occasionally, erythrocytes were observed in the alveolar spaces; however, these appearedto have been dis- placedduring tissue mincing and subsequentprocessing. Therewas IV-17

no evidence that these cells were the result of hemorrhage or an

inflammatory process. Basement membranes and pulmonary interstitium were closely examined. There were some areas in which the back-to-back relationship of the epithelial and endothelial basement membranes could be distinguished. However, there was no evidence of basement membrane separation, degeneration, thickening or edema as a

result of either hydrox or heliox exposure. The general conclusion was that there is no evidence that exposure to hydrox to a depth of 1000 FSW for periods of up to 97 hours results in any temporary or permanent pulmonary damage or in any way interferes mechanically with gas exchange. Figures 20-22 present typical pre-and post-dive electronmicrographs for both heliox and hydrox exposures. It is recognized that this limited number of micrographs cannot present a representative sample of the lung or represent a basis for critical evaluation. More than 300 other micrographs have been

studied as a part of this project. Figure 20

Dog Lung Pre Dive 1000x! Figure 21 Figure 22

Dog Lung Post Dive Hydrox 5000x! Figure 23

Dog Lung Post Dive Heliox 8000x! IV-18

Hematology: Extensive blood studies were conducted using the animal as its

own control. Blood samples were drawn from a foreleg vein. The following determinations were made: 1. Serum alkaline phosphatase S.A.P!

2 ~ Total serum calcium T.S.C.!

3. Creatine phosphokinase C.P.K.!

4. Lactate dehydrogenase L.D.H.!

5. SGOT

6. SGPT

7. Cholesterol

8. Total protein

9. Urea nitrogen

10. Glucose

11. Hematocrit

12. RBC

13 NBC

14. Serum phosphorus Serum Alkaline Phosphatase S.A.P!, Cholesterol and Serum Protein S.P.! were slightly increased while Total SerumCalcium T.S.C.! was decreased in the post-dive samples from both groups of dogs. Creatine Phosphokinase C.P.K.! and Lactate Dehydrogenase L.D.H.! were elevated in the post-dive samples from dogs on 97% Helium: 3% Oxygen and decreased in dogs on 97% Hydrogen: 3% Oxygen. These changeswere not statistically significant and no consistent trend was detected in the other parameters measured Table 7!. IV-19

Increased S.A.P. and decreased T.S.C. raise the possibility of a change in calcium homeostasis in dogs inhaling these gas mixtures. This possibility requires further investigation because of possible effects on neuromuscularirritability and homeostatis. The increased cholesterol seen in both groups of dogs may be due to endogenouslipid mobilization but this is by no meansclear. Elucidation of the underlying mechanismwould necessitate evaluation of fat metabolism and endocrine balance. The increases in C.P.K. and L.D.H. in dogs inhaling heliox may reflect increased muscular activity or subclinical intravascular hemolysis. Increases in C.P.K. and L.D.H. were not seen in dogs on heliox. This mixture may be safer. However, further study is necessary to evaluate this hypothesis. It is gratifying to see that exposure to hydrox does not appear to cause hematological changes which suggest that hydrox is more harmful to the body than is heliox. Indeed, there appears to be no clinically significant difference between the post-hydrox and post-heliox values comparedto their own pre-dive values except for the decrease in the post-dive CPK value for hydrox exposure, and for the quite significant rise in post-dive cholesterol values for both gases. Of special interest is the significant rise in post-dive serum cholesterol levels. This has been reported previously by other workers and seems to be a consistent finding. It would seem to be especially important to determine the source of this cholesterol. If it comes from mobilization of cholesterol deposits it is possible that this could be of benefit to the body, particularly IV-20

if it resulted in reduction of such deposits in the cardiovascular system. Wehasten to note, however,that there is no indication at presentthat this is the case. Onthe other hand,if this rise in circulating cholesterol is the result of new synthesis it mayre- flect an undesirable condition. It could represent an unwanted source of cholesterol and result in further cholestrol deposits- possibly could reflect the loss of precursorsof vitamin D or the steroid hormones. A summaryof blood values may be seen in Tables 7 for those blood enzymes and components of interest.

The Question of Hydrox Inertness in the Body: The large numberof successful hydrox dives to a simulated depth of 1000FSW, and the survival of these animalsfor up to 7 years in apparentlynormal health suggestthat dissolved molecular hydrogeneven at a partial pressureof 23400mmHg 1.1 ATA!does not result in identifiable irreversible physiological damage. On the other hand, there are several indications that molecular hydro- genmay not be completelyinert in the body, particularly at elevated partial pressures. Earlier work by Dole8! indicated that at high partial pressures 00 TORR!molecular hydrogen scavenged alkyl radicals in polyethylene. Further studies by Dole, Wilson, and Fife 9! indicated that hyperbaric molecular hydrogen at an ambientpartial pressureof 8 ATAmay have altered the courseof ultraviolet-induced squamouscell carcinoma in the hairless mouse. Recently, Fife 8! has carried out unpublishedpreliminary studies suggestingthat hyperbaricmolecular hydrogen at an ambientpartial pressureof 8 ATAapparently suppressed the early developmentof IV-21

precancerousskin lesions which appearedwithin 5 days in control mice painted with methylcholanthrene and croton oil. The possible biological effect of molecular hydrogenrelated to the development, prevention or treatment of neoplasmswill not be elaborated upon except to suggest that molecular hydrogen may scavengehighly active alkyl and hydrox free radicals or super- oxides, someof which are a by product of normal cellular metabolism, without disturbing lower energy metabolic activities. However,this would not explain the possibility that the partial pressure of inspired oxygencan decreaseto a lower level in the presenceof helium than in the presence of hydrogen without developing adverse neurological symptoms. If molecular hydrogenenters in mmeway into normal metabolic pathways, it may decrease the efficiency of oxygen metabolism, producing the sameeffect as reducing the partial pressure of cellular oxygen. We found no instance in heliox diving to 1000 FSWin which lowering the partial pressure of oxygen to 166 mmHg.22 ATA! resulted in abnormal EEG's or convulsions. TABLE 7

Pre Dive Post He-02 Post H2-02 MEAN S.D. MEAN S.D. MEAN S.D.

28.00 23.08 30.00 8. 21 Alk. Phos. I .U. 25.36 21.64

36. 80 9.50 21.75 7. 14 SGOT I. U. 33.82 19.42

28. 60 2.70 35.67 25.54 SGPT I.U. 39. 30 25. 99 376. 50 85. 58 92.67 59.05 CPK I.U. 267.00 195.17 427.20 234.89 324.10 477.59 LDH I.U. 319.36 288.58 Cholesterol MG/DL 175.67 43.08 207.20 21.38 210.50 58.26

1.23 7.52 1. 80 Total Prot. GM/DL 6.97 1. 38 7.60

2.17 13. 63 3. 77 UREA N MG/DL 13.92 4. 54 12.14

0.29 0.98 0.38 Creatinine MG/DL 0.93 0. 32 0.98

1.41 9.31 1. 18 Ca MG/DL 10.53 0.72 9.98

0.51 5.43 1. 45 P MG/DL 5.45 1.07 4.68 102.40 16.82 55.50 31.30 Glucose MG/DL 82.73 14.14 F 00 7. 36 43. 75 6.18 39. 80 HCT 43.57 SPECIAL TECHNICAL PROBLEMS AND THEIR SOLUTIONS

HYPERBARIC CHAMBER; SPECIAL PROBLEMSAND THEIR SOLUTIONS: Manyhyperbaric chambersnow in operation could be madesuitable for hydrox diving without major modifications. Any chamberwhich is safe for use with oxygenshould be even safer for use with 3%02 in hydrogen, provided care is exercised to avoid leaks. In fact, we believe that flooding a chamber with oxygen-enriched gas is far more hazardous than flooding the chamber with hydrogen. This is perhaps best illustrated by the work of Doerr and Schreiner 3! who showed that even compressed air under pressure could sustain a vigorous fire. On the other hand, as already mentioned, a gas mixture containing 3%oxygen and 97%hydrogen could not be ignited even with an open spark. Great care must be taken to assure that hydrogen gas will not leak out of the chamber into the laboratory. While such leaks may occur at any penetration, door or port, our experience has shown that leaks are more likely to occur at electrical penetrations and on large diameter hose or pipe fittings. Potting techniques described in Appendix 5 have been successfully used in our laboratory for several years with little trouble. Most of our penetrations are tested at a pressure of 1800 psi with helium and are completely leak proof. Someof these have been in constant use for over three years without failure. If hydrogen is to be used in the chamber it is essential that the chamber and all ancillary components be tested with helium to at least 10% over the proposed working pressure before the dive begins. Many leaks not apparent with nitrogen will become major problems with helium or hydrogen at the same pressure. Rarely will V-2

a leak appear with hydrogen if it did not appear during a check

with helium. All fittings, valves, regulators, doors and ports are checked for leaks every 100 feet during compression and at least once during each subsequent4 hours. In addition, an electronic hydrogen warning system is in constant operation in the chamberarea. An additional precaution has been taken by installing a closeoff valve at every chamberpenetration except for the electrical penetration. In this way hoses, gauges, etc. can be isolated and replaced in event of failure without aborting a dive. Although electrical penetrations do not have close-off valves installed, the penetration is constructed in such a way as to permit wires to be cut and the penetration capped in event a leak occurs.

CARBON DIOXIDE REMOVAL Any chamberdive which is to last more than a few minutes must haveprovisions for C02removal. This maybe accomplishedeither by an internal or external scrubber. Oneunit in commonuse in the diving industryis the Lindberg-HammarModel M-9.* This unit employsa sealed explosion-proof 24-volt AC/DCmotor with magnetic coupling to the fan. It also contains a basket in which any of the usual granularC02 absorbers may be used. This unit maybe placed inside the chamber and be controlled with a variable auto transformer outside the chamber. Another satisfactory motor is manufactured

by Rotron. We also have built an external scrubber, using the same type of Lindberg-Hammar motor and basket. This was accomplished by construct- ing a four foot long container from high pressure oil-field pipe.

* Lindberg-Hammar Inc. Fort Worth, Texas

** Rotron Inc. New York, New York V-3

Themotor was permanently mounted in the bottomsection while the scrubberbasket may be insertedor removedfrom the top. Oneinch diameterhigh pressurehoses connected the unit to the chamber. CO2scrubbers for small animal chambers areeasily made in the laboratory. Theyusually consist of a can5-6 inchesin diameter with a screenat eachend. Thefan is driven by an inductiontype motorbuilt for continuousoperation. Thecondenser for the motor is remotedto the outside of the chamber. In this waythere are no comutatorbrushes inside the chamberand no external pressure is placedon the condenserwhich may be oil filled. Thesescrubbers haveoperated successfully in our laboratoryfor over 5,000hours of hydrox exposure. It is desirableto use slow-blowfuses in the motorcircuit. Thesefuses should be rated slightly belowthe lockedrotor current

flow of that particular motor. Considerationshould be given to the C02absorbent used. We choseSoda-Sorb since its CO absorbingcapacity is greater than 2 someother types. Initially weobtained this fromeither a medical or scientific supplydistributor. However,after a long series of problemsit wasdiscovered that theremay be a greatvariation in CO2absorbing capacity between batches of this absorber.Further, it apparentlyis possiblefor scrubbermaterial to meetmedical gradespecifications and still not haveenough CO2 absorbing capacityto satisfy our ownneeds. As a result, arrangements weremade with a manufacturingcompany to producea supply of fresh Soda-Sorbhaving a CO capacity at least 4 times better 2 than the usual medicalgrade. In someinstances 10 poundsof ab- sorberwould adequately remove C02 produced by a 60 pounddog for V-4

up to 3 days. These problems would suggest that larger users of C02absorbers may find it wise to deal directly with a factory producing absorbent material. Soda-Sorb especially manufactured for use in diving now is available.*

Heat Loss: As discussed elsewhere, it is extremely important to maintain ambient chamber temperature in the range of 33'C to 35'C when hydrox is the diving gas. Further, since the temperature tolerance of the subject seems to be narrowed in this environment it is highly desirable to allow the subject to make his own fine adjustments in his temperature exposure. This can be accomplished by heating the chamber from one side wall only, using an external heat source. In one of our larger chambers 0 cu ft! a 220 v thermostatically controlled electrical space heater is placed 6 inches from the outside wall of the chamber. In small chambers an infrared bulb or a 110 v heating strip, thermostatically controlled may be employed.

Ammonia Control:

It is essential the urine be removed from the chamber as quickly as possible since the high temperature of the chamber greatly accelerates breakdown of urine into ammonia. It is not clear how serious the ammonia build-up is on the subject. On an early dive of 54 days duration no effort was made to remove urine.

As a result, ammonia levels became high enough to bother the technician when the dog was removed from the chamber at the ter- mination of the dive. The dog showed no gross evidence of pulmonary damage, and no subsequent respiratory problems appeared. On the other hand, intuitively it would seem that ammonia build-up

*Soda-Sorb H.P. Dewey-Almy Chemical Div., Atlanta Ga. V-5

should be avoided. Studies on the effects of ammonia on the body under pressure have been much too brief to draw valid conclusions since in the past considerable precautions have been taken to avoid ammonia build-up in the chamber. In the larger chambers this removal is accomplished by flushing water over the chamber

subfloor. In the smaller mouse and rat chambers absorbent flooring is used which is changed periodically by bringing the chamber to ground level. These small animals can be brought to the surface and returned to depth in 3-6 hours, depending upon the depth of exposure. SAFETY CONSIDERATIONS

Combustibilit of H drox: Of primaryconcern is the questionrelated to the combustibility of hydrogen-oxygengas mixtures. Thefirst concernover this matter appearsto havebeen raised by Caseand Haldane! who apparentlyconsulted Professor A.G.C. Egerton, Secretary of the RoyalSociety, beforethey undertooktheir studyof hydroxin 1941. Professor Egerton assured them that a mixture containing 4%oxygen and 96%hydrogen was entirely safe from explosion, as was a mixture containing 68.7%hydrogen and 31.1% air. More recent and carefully documentedwork by Dorr and Schriener 3! place the safety limit at 5.2%oxygen and 94.8% hydrogen. There has beenword that someRussian work suggests the safety limit is somewhatlower than that identified by Dorr and Schreiner but this report cannot yet be confirmed. In our laboratory we have considered the work by Dorr and Schreiner to be reliable. However, we, like most other workers have arbitrarily limited oxygen levels to about 3%. Wehave an absolute maximumpermissible oxygenlevel of 3.5%to allow for slight variations in commercially produced hydrox. However, our ownmixing systemmakes is possible to hold oxygenlevels to

3% + 0.05%. The 3%oxygen limitation posesno serious problemto the life of the diver who is deeper than 200 FSWsince it represents a normaloxygen level at that depth ata!. As the diver goesdeep- er he maywish to lower his inspired oxygento less than 3%. The rationale behind this is further discussed in Appendix1 and 2. Perhapsthe best indication of the safety with which hydroxcan be used is to consider a safety record. Wehave carried out in excess VI-2

of 6,061 hours of animal and human exposure to hydrogen-oxygen

mixtures in chambers over the past 10 years without accident. These range from simulated dives to depths of 200 feet of sea water FSW! to 1000 FSW. Total times of individual dives range from 5 hours to over 5 days in duration. It should be noted that in all but about 20 hours of hydrox diving we have elected to flood the chamber with the hydrox mixture. So far, in the human dives the hydrox has been delivered by mask with overboard dumping of exhaled gas. On the human dives the chambers have contained 160 mmp02 at all depths with the balance nitrogen.

H dro en Embrittlement: The matter of hydrogen embrittlement has become of increasing concern as interest in hydrogen for fuel and other uses has increased. Perhaps the greatest concern to divers came with reports by Michaud 4! and his associates who noted that door failure occurred in one of their animal chambers, apparently due to hydrogen embrittlement.

A considerable amount of work has been carried out to study this problem. There is no doubt that such embrittlement does, indeed, occur, and that it varies with the type of metal and alloy employed. On the other hand, many of these embrittlement studies were conducted using extremely pure grades of hydrogen. This becomes of special importance in view of the work by Kesterson 5!, who showed that in some alloys, the embrittlement was reduced in direct proportion to the increased partial pressure of oxygen in the hydrogen gas. If these data are replotted and extrapolated to bring the oxygen levels within the range used in diving, it can VI-3

be seen that within the range of the data points experimentally derived the reduction in embrittlement appears to be significant if oxygenis presentin the hydrogen.For example,in onetype of stainless steel the data show that when hydrogen contains 150 ppm .015%of oxygen!there is only a 7.5%reduction in strength. If these data points are graphedit can be seen that the reduction in strength is a semi-logarithmic function of the oxygen con- tent. Thus, by extrapolation, whenoxygen levels reach 500 ppm .05%!it wouldappear that the strength wouldbe reducedby only 0.1%from its original strength. Since the metal tested also had been in irradiated it. is possible that non-irradiated metal may

suffer even less loss of strength. It is clear that different alloys respond differently to hydrogenembrittlement. Indeed, the abovecited metal CW301 stainless! showedmuch greater loss of strength than the other

3 alloys tested at that time. It should be realized that the hydrogen in diving chambers always contains a far greater amountof oxygenthan 500 ppm. In fact, at a depth of 4,000 FSWthe oxygen levels probably would be about 0.4% 80 mmHg!. While this observation does not provide a totally satisfactory answerto the concernover hydrogenembri"tlement it would suggestthat evenwith the reduction of strength from embrittlement a chamber still may be well within the design toler;-..nce of the device if high quality steel has been used.

Flash-back Arrestor: Since the hydrox chamber will be venting hydrox almost constantly this gas must be exhausted in a safe manner. It is not VI-4 enough simply to vent the exhaust to the outside of the building. While 3% hydrox is not in itself explosive, at low flow rates air will move retrograde through the exhaust line, bringing oxygen tothe exhaust valve of the chamber. This would create an explosive mixture in the exhaust line. This problem easily can be overcome by directing the exhaust line to the bottom of a steel drum filled with water. Exhaust gas then must bubble up through the water, thus effectively isolating the hydrox line both from the back flow of air and from any flame which might develop above the drum. The pipe leading into the water drum must, of course, be metal for at least the last 10 feet of its length. Gas analysis exhaust tubing may be handled in two ways. One is to place an aquarium air stone on its end and place it 4" below the surface of a water-filled metal container. The purpose of the air stone is to eliminate varying back pressure which is created by large bubbles. This may affect the analyzers. The other method would be to place a mine safety screen over its end. This would prevent the flash-back of a flame but would not prevent retrograde diffusion of air when the flow rate is slow. APPENDIX 1

Protocol for H drox Diving with Animals: The following detailed procedure is used for hydrox chamber diving with animals: 1. Complete the pre-dive check list 2. Close chamber door and begin compression to 200 FSWwith

pure helium or nitrogen. Comments: At the start of compression the chamber contains 21%oxygen at a partial pressure of 160 mmHg. Uponreaching 200 FSW atmospheresabsolute ata!! the chamber still contains approximately 160 mmHg of 0 , reducedby the amount.metabolized. However, this oxygen now represents only 3% of the total gas. As previously indicated, since a gas mixture containing less than 5%oxygen and 95%hydrogen cannot be ignited by a spark, hydrogen now may safely be introduced into the chamber. Further, since at 7 ata a gas mixture containing 3%oxygen still represents 160 mm Hg partial pressure of oxygen, the subject is exposedto normal life-sustaining oxygen levels. 3. Check all fittings and ports for leaks. Comments: If the chamber is never allowed to contain oxygen levels above 5%, there is no danger of explosion or fire inside the chamber. However, leaks may be serious since such leaks will allow hydrogen to be mixed with the 21%oxygen of the air. Pro- tection against chamber, cylinder or regulator leaks, therefore, is essential. Thus, leak checks are made before introduction of hydrogen, and after reaching each additional 100 feet of depth.

If leaks are discovered further compression is prohibited until VII-2

they can be stopped. 4. Flush chamber with 3% hydrox % oxygen in 97% hydrogen!. Comments: This flush is designed to eliminate the helium and nitrogen gases which are present upon reaching 200 FSW. The amountof gas required to carry out this gas exchangedepends upon the size of the chamber, the amountof residual nitrogen or helium which can be tolerated, and the efficiency of gas mixing in the chamber. Practice has shown that for a 30 cu ft chamber, approximately 1400cu ft of hydrox is required to reduce the other inert gases to 0.5%. This figure would, of course, vary with individual facilities. It should be noted that since the chamber contained 3% oxygen before flushing began, and that the flushing gas also contained 3%oxygen, the diver would see no change in oxygenation. 5. Compress to the desired depth with pure hydrogen. Pause each 100 FSW for leak checks. Comment: It is important to keep the chamber oxygen levels within a normal physiological range. If no additional oxygen is added, the partial pressure of oxygen remains at about 160 mmHg. Since it is gradually reduced by metabolic requirements of the diver, if compression is quite slow it may be necessary to add more oxygen. This must. be done in the manner described in paragraph 6 below. The rate of compressionmust be carefully controlled as the dive goes deeper due to the possibility that the diver will develop the high pressure nervous syndrome HPNS!. For dogs, a pause of 30 minutes each 100 feet for leak checks appears to allow a dive to 1000 FSW on hydrox without the presence of HPNS. VII-3

6. Oxygenreplacement. This is accomplishedby introducing a hydrox mixture containing 3% oxygen. Comment: It is essential to avoid introducing an explosive oxygenmixture into the chamber. Useof already mixedhydrox will allow the input of 3%oxygen with the balance being hydrogen. Since dives deeper than 200 FSWwill call for less than a 3% oxygenlevel in the chamber,adequate oxygen levels canbe maintained by addingthe 3%mixture. It usually is possible to establish a steadybleed-in of hydroxto just balanceoxygen utilization while the dive is in progress. If the oxygen level must be raised during compression it may be accomplishedby carrying out a part of the compressionwith hydroxin place of pure hydrogenas describedin paragraph5 above. Whenthe oxygenconcentration again reaches the desired level, compression is continued with pure hydrogen.

1. Begin ascent by opening the exhaust valve. Bleed in hydrox as required to raise the oxygen level. Comment: It is necessary not only to follow carefully the planneddecompression table, but to maintainadequate oxygenation as well. If the depth is below 200 FSW,the oxygen levels will have been held at less than 3%. For example, at 1000 FSWthe oxygenlevel in the chambershould be betweenO.o% and 1.5%.! By the time the diver ascendsto 200 FSWthe oxygenlevel must be returned to 3%. This is accomplished by flushing in 3%hydrox. As the diver approaches200 FSW,the quantity of hydrox utilized in the flushing becomes increasingly greater. 2. Upon reaching 200 FSW, begin flushing with 3% oxygen-

97% helium heliox!. Comment: The flushing with 3% heliox will assure that at no time hydrogen will be present with more than 3%oxygen. This flushing at 200 FSWalso assures adequatetissue oxygenation since as indicated above, 3% oxygen provides 160 mm Hg partial pressure at 7 ata. Flushing is continued until the hydrogen level is reduced to below 3%, at which time it no longer is an explosive mixture with air. 3. Introduce the desired oxygen or nitrogen-oxygen nitrox! mixture required by the decompression profile. Comment: If it. is desired to decompress on heliox, it is only necessaryto bleed-in sufficient oxygento raise the chamber oxygencontent to the desired levels as decompressionproceeds. If compressedair or other nitrox mixture is desired, it is necessary to flush with that mixture to eliminate helium. Consideration should be given to possible oxygen toxicity if an oxygen-enriched environment is contemplated. After elimination of the hydrox mixture from the chamber, de- compressionmay proceed in accordancewith usual chamberdecompres- sion practices. VII-5

APPENDIX 2

Protocol for Hydrox Chamber Divin with Humans:

Compression: 1. Proceed through both the 24-hour pre-dive checklist and

the "Day of Dive" checklist. 2. Diver enters and door is secured. Compression begins to

100 FSW using pure nitrogen. Comment: The diver may breathe chamber air during this de- scent, or may breathe compressed air by mask, depending on the table used. This compression begins to reduce the percentage of oxygen within the chamber without reducing its partial pressure. Although there is an increased percentage of nitrogen in the inspired gas, it is not sufficient to create unacceptable nitrogen. narcosis. The extra nitrogen uptake must be considered during subsequent decompression. 3. Diver begins to breathe 10% oxygen-90% helium by mask, and compression toward 200 FSWis resumed using pure nitrogen. Exhaled gas is exhausted overboard into a flashback arrestor. Comment: Although 10% oxygen is slightly hyperoxic to the diver, it does not create a toxicity problem during the relatively short time it is used. Further, this mixture is standard in mixed gas diving. Compressionof the chamber further reduced the percentage of ambient oxygen so that upon reaching 200 FSW ata!, the chamber atmosphere is only 3% oxygen. This will not create an explosive mixture in event hydrogen gas subsequently leaks into the chamber. At the same time, in event of mask failure the diver can breathe the chamber atmosphere without danger to life. Of course, the chamber cannot be flushed with air to remove a build-up of hydrogen. Further, flushing with pure nitrogen would remove excess hydrogen, but. VII-6 would lower oxygen levels. The chamber atmosphere then no longer would support life in case of mask failure. H2 leaks into the chamber should be avoided. 4. Shift breathing mixture to the 3% heliox manifold for two minutes. Comment: This is done instantly by closing the 10% heliox line and opening the 3% heliox line between breaths. Since the intake line is cleared of the previous gas in 3-4 breaths, the 1- minute breathing period provides an adequate safety factor to purge the intake line and regulator. Since the exhaled gas expands upon leaving the chamber, the exhaust line to the flashback arrestor completely is cleared in two breaths. This assures that the exhaust line also will not contain more than 3% oxygen by the end of the 2-minute period. This 2-minute period also allows time for the chamber attendants to shift gas lines to the hydrox manifold. 5. Shift breathing mixture to the 3% hydrox manifold. Comment: The diver now is breathing a 3% hydrox mixture. Overboard dump of exhaled air assures that there will not be a buildup of hydrogen in the chamber. 6. Compress to desired bottom depth with pure nitrogen. Comment: The diver may remain on the 3% hydrox breathing mixture for an indefinite period down to a depth of 475 FSW. Based on Lamberstsen's work, he could go much deeper for shorter periods of time without being in danger of serious oxygentoxicity. Indeed, our work suggests that he may be able to remain for a week or more at a depth of 625 FSWwhile breathing 3% hydrox. If there is concern over the effects of oxygen at greater depths, the breathing mixture may be shifted to one containing less oxygen. VII-7

1. Ascend to 200 FSW by reducing chamber ambient pressure in

accordance with the decompression schedule.

Comment: The diver remains on the mask during this phase. His breathing mixture should be shifted back to 3%hydrox if it had been changed to a lower oxygen percentage during the dive. If the chamber had been compressed with pure nitrogen to a depth significantly deeperthan 200FSW, the atmospherewill contain less than 3% oxygen. As a safety precaution 3% oxygen in nitrogen should be flushed in to bring oxygen levels to about 3%. In this way it will be assured that the diver will have adequate oxygen in event of mask failure. Chamber oxygen levels should not be raised above 3% as long as the diver is breathing hydrogen. 2. Shift breathing gas to 3% heliox for 2 minutes. Comment: This will assure that hydrox is flushed out of the entire breathing system. The question of isobaric bends may be raised since there is an almost instantaneous shift from hydrogen to helium. This problem has not been observed in our work. 3. Check chamber gas to be sure it contains no more than 3% hydrogen.

4 . Shift breathing gas to 10% heliox. Comment: This will flush out the 3% heliox and make it possible to bring the diver to within 33 FSWof the surface without danger of hypoxia. 5 . Decompress in accordance with standard techniques. Comment: Examples of both animal and human decompression tables developed in this laboratory are presented. They are discussed in more detail elsewhere' VII-8

APPENDIX 3

Lungbiopsies on dogs maybe carried out in a manneralmost identical to that used on humans. The technique has been outlined elsewhere but is briefly described as follows: The dog is tranquilized with Acepromazineadministered inter- muscularly. The usual dosageis 1 mg per kg of body weight. The right chest area is shavedand cleansedas for other surgery and a location is selected, exercising caution to avoid the heart. This may be done by bringing the elbow of the foreleg back to mid-chest. The angle of the elbow is the approximate location of the heart. A dependent lobe will be entered by staying at least 4 cm caudal to this angle. An area over the center of a rib is selected as the point of penetration and is infiltrated with Xylocaneor other local anesthetic agent, following which a 2 or 3 mmincision is madeperpendicularly through the skin directly over the rib with a stylet-type blade. The incision should penetrate to the lateral surface of the rib. A sterile Vim True Cut biopsy needle then is inserted perpendicularly through the incision to the surface of the rib. Upon contacting the rib the needle is slid cephaladwhile still held perpendicular to the skin. When the tip of the needlecan be felt to slide over the anterior border of the rib it is forced directly inward, penetrating the parietal pleura. Care must be exercised to keep the needle as close as possible to the rib border to avoid penetration of the intercostal vessels. Whenthe tip of the needle has penetrated the parietal pleura it will rest on the visceral pleura covering the lungs. VII-9

This can be confirmed by the tendency of the back of the needle to movein a cephalad or caudal direction in synchronization with respiration. At the momentof greatest exhalation the coreing needle is quickly advancedinto the lung tissue, rapidly followed by the outer sleeve which cuts the biopsy plug. Theneedle then is quickly withdrawn. Since the point of penetration through the skin slides caudally a few millimeters it closes and seals the needle pathway. This tends to reduce the possibility of pneumo- thorax. The animal recovers in a few minutes and resumes normal activity. Lungbiopsy tissues then maybe transferred to glutaraldehyde fixative and preparedin a standardmanner for electron microscopic study. VII-10

Appendix 4

Protocol for Liver Biops This biopsy may be done with a Menghini needle or with a Vim True Cut needle. Our technique employsa modified Menghini needle of our own design. 1. The animal is tranquilized by acepromazineas described above' 2. The chosen area, usually subcostal, is shaved, sterilized and infiltrated with Xylocaine to the depth of the peritoneum. 3 ~ An incision through the skin to the peritoneum is made with a stylet. 4. The Menghini needle is attached to a 6cc syringe containing 3cc of sterile saline. The needle then is advanced through the incision so that the tip rests on the surface of the liver capsule. 5. Approximatelylcc of saline is ejected to clear the needle.

It then is pressed against the capsule. 6. The plunger of the syringe is withdrawn a few cm to produce a vacuum and seal the liver to the needle tip. While holding the vacuum, the needle is quickly advanced to the desired depth and without pause quickly and completely withdrawn. Upon leaving the body, the vacuumdraws the biopsy tissue into the saline solution in the syringe ~ The needle then can be removed and the tissue expressed into the fixative. Appendix 5

Potting Techniques: The potting of electrical penetrations for the C02scrubber becomescritical since the developmentof even a small leak in sucha fitting often results in the requirementsto abort the dive. A numberof epoxy-type electrical potting materials were tried without success. The potting material finally settled on was Scotchcast 45. This material will withstand the stress of repeated compression even in a hydrogen environment. The potting process is as follows: l. Insulation is removedfrom the multistranded wires for a distance of 1-2 inches at the point where they will pass through the fitting. The strands are solderedtogether to makea single solid bundle. The area then is coated with the potting compound

to provide a good bonding. 2. The inside of the fitting is cleaned on tne inside with solvent and scraped to the bare fresh metal with a circular file.

It then is coated with the potting compound. 3. Theepoxy-coated wires are passedthrough the fitting and a plug of clay or gumrubber is moldedaround one end to provide

a seal. 4. Additional potting solution is poured in the open end, and the entire unit is placed in an ovenat a temperatureof 140'F. Care is exercised to assure that the wires are not touching each other or the fitting wall. The fitting is allowed to cure for 24 hours undisturbed. It then may be removed, cooled, and tested. Penetrations ranging from >4"to 1" may be prepared in this

way. BIBLIOGRAPHY

1. Lavoisier, A. L. Premier memoiresur la respiration des animaux. Memoisede 1'Academic des Science, p. 185, 1789.

2. Case, E. M., Haldane, J. B. S. Humanphysiology under high pressure. J. Hygiene41:225-249,

1941.

3. Zetterstrom, Arne Deep-seadiving with synthetic gas mixtures. Mil. Surgeon

103:104-106, 1948. 4. Bjurstedt, H., Severin, G. Theprevention of decompressionsickness and nitrogen narcosis by the useof H2as a substitutefor nitrogen. Mil. Surgeon

103:107-109, 1948.

5. Personal Communication

Bjurstedt, H., 1976. 6. Brauer, R. W., Johnson, D. O., Pessotti, R. Effects of hydrogenand helium at pressures to 67 atmospheres on mice and monkeys. Fed. Proc. 25:202, 1966.

7. Brauer, R. W., Way, R. O., Perry, R. Separationof anaestheticand convulsant effects in mice breathingHe and H2 containing atmospheres at 50 to 150atm. Fed.

Proc. 26:720, 1967. 8. Brauer, R. W., Way, R. 0., Fructus, X. A generalizedmethod for determiningpotency of hydrogenin man. Proc. Int. Cong. Physiol. Sc. VII, 1968. 9. Brauer, R. W., Jordan, M. R., Way, R. 0. Modification of the convulsive seizure phase of high pressure excitability. Fed. Proc. 27:254, 1968. 10. Brauer, R. W., Jordan, M. R., Way, R. O., Sherman,M. E. High pressure hyperexcitability syndromein the Squirrel Monkey. Fed. Proc. 28:655, 1969.

11. Brauer, R. W., Way, R. 0. Relative narcotic potenciesof H2, He, N2, and their mixtures. J. Appl. Physiol. 29:23-31, 1970. 12. Brauer, R. W., Way, R. 0., Jordan, M. R., Parrish, D. E. Experimental studies on the high pressure excitability syndrome in various mammalianspecies. In. Proc. of the IV of the Symp.on UnderwaterPhysiol. Phila. Penn.Ed. Lambertsenpp. 545-

550, 1971 '

13. Bennett, P. B. Performance impairment in deep diving due to nitrogen, helium neon, oxygen. In. Proc. Third Symp.Underwater Physiol. Ed. Lam- bertsen pp. 327-340. Baltimore, Williams and Wilkins.

14. Cullen, S. C ~ , Gross, E. G. The anesthetic properties of Xenon in animals and humanbeings with additional observations on Krypton. Science 113:580-582, 1951. 15. Edel, P. O., Holland, J. M., Fischer, C. L., Fife, W. P. Preliminary studies of hydrogen-oxygenbreathing mixtures for deepsea diving. TheWorking Diver, SymposiumProceedings, Feb. 1972, Columbus,Ohio, Washington,D. C., Marine Technol. Soc. J. pp. 257-270. 16. Michaud, A., Pare, J., Barthelemy, L., Le Chuiton, J., Corriol,

J., Chouteau, J., Le Boucher, F. Premieres donnes sur un limitation de 1'utilization du milange oxygen-hydrogenepour la plongee profonde a saturation. C.R.

Acad. Sc. Paris, 269:497-499, 1969.

17. Lazarev, N. V. BiologicheskoyeDeistviye Gazovpod Davleniyem Biological Action of Gases Under Pressure!. Naval Medical AcademyPress.

Leningrad. 1941. 18. U.S. Helium may be gone in 30 years. The Oil and Gas Journal.

June 8, 1970.

19. Dougherty, J. H. Jr. The use of hydrogen as an inert gas during diving: Pulmonary functions during hydrogen-oxygenbreathing at pressures equivalent to 200 feet of sea water. Bureau of Medicine and Surgery, Navy Dept. Report Number 801, 1974.

20. Lambertsen, C. J ~ Collaborative investigation of limits of human tolerance to pressurization with helium, neon, nitrogen. Simulationof density equivalent to helium oxygen respiration at depthsto 2000, 3000, 4000 and 5000 feet of sea water. In proceedings of the fifth Symposiumon UnderwaterPhysiology Ed. Lambertsen,Publication Press,

Inc. Baltimore. 1976. 21. Fife, W. P., Mezzino, M. J., Naylor, R. Thedevelopment and operational validation of accelerateddecompression tables. Sixth Symposiumon Underwater Physiology 6-10 July 1975,

San Diego, Cal. In Press! ~

22. Webb, P. Body heat loss in undersea gaseous environments aerospace medicine. 41:1282-1288, 1970. 23. Doerr, V. A., Schreiner, H. R. Second SummaryReport on cimbustion safety in diving atmospheres. Defense Documentation No. AD 689545, Government Printing Office, Washington, D. C., May 1969.! 24. Michoud, A., Le Chuiton, J., Pare, J., Barthelemy, L., Balouet,

G., Girin, E., Corriol, J., Chouteau, J. Bilan d'une experimentation animale de plongees anx melanges hydrogene-oxygene.Marine Nationale Groupd'Etudies et Recherches

Sous-marine, Loulon, 1973.

25. Kesterson, R. L. Effects of irradiation and oxygen on hydrogen environment embrittlement of selected alloys. Hydrogen Embrittlement Testing ASTM-STP 543, American Society for Testing Materials, 1974, pp.

254-263.

26. Klemm, W. R. Electroencephalography In: Applied Electronics for Veterinary Medicine and Animal Physiology. Ed Klemm, Charles C. Thomas,

Pub. Springfield, Ill.

27. Hills, B. A. Vital issues in computing decompression schedules from fundamentals. II Diffusion versus blood perfusion in controlling blood: tissue exchange. Int. J. Biometeor. 14:323-342, 1970.

28. Dole, Malcolm Advances in Radiation Chemistry. M. Burton and J. L. Magee, eds. Wiley, New York. 4:307, 1974. 29. Dole, Malcolm, Wilson, F.R., and Fife, W. P. Hyperbaric HydrogenTherapy: A possible treatment for cancer.

Science 190:152-154, 1976.